We propose to explore ways to use AAV-based gene delivery to restore and/or maintain retinal cone structure, function and vision-dependent behavior. Our overall hypothesis is that appropriately designed, gene-based therapies following intravitreal (IV) injection will be clinically useful for a wide variety of retinal diseases with cne dysfunctions. The need for developing the IV injection option is based on recently published results for 15 patients from the current NEI sponsored LCA2 clinical trial (NCT00481546). Single or multiple subretinal (SR) AAV vector injections resulting in vector blebs that detached the fovea resulted in little or no vision gain for the patient and in some a slight loss in visual acuiy. In contrast, equivalent subretinal vector doses that did not detach the fovea mediated substantial and persistent gains in vision. Therefore, developing alternative ways to more safely transduce foveal cones is the focus of this proposal. To that end, we will test the IV route of vector injection and employ novel AAV variants that we have just developed in an effort to validate and optimize the ability to correct cone defects in a mouse model of achromatopsia 2 without detaching the retina. For this investigation, we will use a mouse model of human achromatopsia with CNGA3 mutations using several novel tyrosine to phenylalanine plus threonine to valine mutant capsid AAV vectors as the gene delivery vehicles for intravitreal injection. Although IV injection of vector can transduce a large area of the inner retina without causing retinal detachment, conventional vectors in the vitreous cannot penetrate the inner retina and transduce rod or cone cells in outer retina. Developing and validating a retina-penetrating property for AAV vectors is therefore a key unmet need in the field and a challenge for current AAV-mediated gene replacement therapy. Another pivotal element of this project is the selection of an animal model that not only possesses a genetically and phenotypically well-characterized cone degeneration, but also has been established to respond well to SR gene-based therapy so that screening these novel AAV vectors for functional and structural cone rescue from the vitreous can be carried out with confidence with reference to an established therapeutic standard. We will therefore focus on a naturally occurring mouse model of human achromatopsia 2 (recessive CNGA3 mutations), the Cpfl5 mouse, that exhibits cone dysfunction/degeneration similar to that seen in humans and responds well to SR AAV vector gene therapy. Therapeutic success in this animal model will provide a template for safer IV treatment trials not only for human achromatopsia but also for other human retinal diseases affecting foveal function such as late stage retinitis pigmentosa, cone and cone/rod dystrophy and potentially macular degeneration.